The Handler's Tuesday

The alarm is set for 5:42, because 5:45 is when Karen’s alarm goes off, and the three-minute lead lets me start the coffee without the bedroom door opening on the smell of it. She does not like to be woken by coffee. She has not liked it since we were married, and I have not changed the system since 2003.

I take the coffee to the back deck and stand in the dark and drink half of it before the dog comes out. It is November and the deck boards are cold through the slippers. Across the cul-de-sac the Petersen kitchen light comes on, which is Mr. Petersen, who is a retired urologist and rises at six. None of my neighbors know what I do. The cover story is “logistics consulting, federal contracts,” which is the kind of thing that ends a conversation at a barbecue inside two questions, which is its design.

I make Maddy’s lunch. Turkey on the soft white bread she still prefers at fourteen, an apple cut and lemon-juiced so it doesn’t brown, two of the small bags of pretzels she likes. I write a note on a Post-it that says good luck on the chem quiz and put it in with the apple. I have written some version of this note approximately three thousand times. Maddy is the second daughter; the older one, Elise, is at Wesleyan and writes me long emails about her thesis on Hannah Arendt, and I read them carefully and answer them in two paragraphs, the second of which is always about her sister.

I leave the house at 6:48.

The commute is forty-one minutes when the Beltway is reasonable. I take 267 east and exit at the second of the office parks past the airport. The sign at the exit reads DULLES TECHNOLOGY CORRIDOR — EXITS 9–11, and beneath the highway department’s sign someone, twenty years ago, mounted a small carved owl on the post. It is weathered now. Three of the wing-feather grooves have filled with lichen. I noticed it within my first week on the desk, in 2001, and I noticed that I noticed it, and I made a private decision not to look at it again, and the decision held for about eight years and then it was simply furniture and I stopped seeing it the way you stop seeing the back of your own hand. It is still there. I do not look at it this morning either, but I know it is there, in the way you know the molars are at the back of your jaw.

The facility is the third building from the entrance in a row of six identical four-story buildings clad in beige composite panel. The door plate reads LOGISTICS SUPPORT GROUP — SUITE 312. The lobby contains a sign-in tablet, a vending machine with bad coffee, and a security desk staffed by a contractor named Eddie who is reading a paperback about the Marine sniper who killed Yamamoto. Eddie waves. I wave. I take the elevator to 3.

Suite 312 is what the cover represents itself to be. It contains seven cubicles, a copier, a coffee maker, a small conference room with a whiteboard. Three people work there. They process actual freight-routing optimization contracts for two actual federal customers. The contracts are real and the work is competent. The office turns a small profit on its cover business, which is one of the things the AARO reorganization tightened up in 2022, on the reasonable theory that an office that loses money is more interesting to an inspector general than an office that does not.

The door to my actual office is in the back corner, marked ARCHIVE — PLEASE SEE J. ROBERTS. The keypad is mounted inside the doorframe and is not visible from the suite. I enter the four-digit code and then the second six-digit code, and the inner door opens onto a small landing and a stair down half a flight to the cleared spaces, which occupy the basement and sub-basement of the building and extend, by a covered corridor, into the basement and sub-basement of building four. The cleared spaces are larger than the cover suite by a factor of about eleven. They are also, mercifully, not visible from any window I will see today.

I am at my desk at 7:34. The queue is in the inbox and on the secure-side terminal and in the three folders the night-watch officer has stacked, neatly, on the left side of the blotter.

The queue today is, in priority order: a P5 maintenance cycle to confirm for tonight’s window in the Shenandoah box; an incident report from a sub-handler named Pruitt out of the Fort Detrick subordinate desk; an inbound signal from OP-CHANNEL-3 timestamped 04:11 local that wants upward routing before the noon liaison call; a media-watch flag from Section 14 about a reporter at one of the longer-form outlets who has been making freedom-of-information requests at the wrong drop site; and a meeting at 14:30 at the Reston interface station for a quarterly observation. Five items. A normal Tuesday.

I clear the maintenance confirmation first because it is the cheapest. The work-order is a routine P5 on a substrate the registry has rotated three previous times, no SRE flags, the attending team is one I’ve cleared four times this fiscal year. I review the AMA-2 weather window — clear, no civil overflight, the National Guard exercise that briefly clouded last month’s calendar is finished — and I sign the confirm. The system stamps it with my handle and routes it to Section 11. Time elapsed: nine minutes.

Pruitt’s incident report takes longer. Pruitt is thirty-one and has been on the Fort Detrick desk for two years, and he is competent but writes the way young careerists write, which is to say he buries the actual issue under a page of throat-clearing. The actual issue, when I find it, is that one of his subordinate operators — a contracted intake nurse, civilian, has been with the program eleven years — failed to file the post-event observation log within the 24-hour window required under P9. She filed it at 31 hours. The substrate is fine. The redaction held. The lapse is administrative. Pruitt has written it up as if it might warrant a Section 11 review.

I call Pruitt on the secure line. It’s a 7-day reissue of the SOP reminder, I tell him. Not an escalation. The nurse’s name does not appear in the upward report. You log it in your desk file, you note the corrective action, you move on. If she does it again inside the calendar year, then we have something to talk about. Until then, you are creating paperwork that will live somewhere and that we will both regret. Pruitt thanks me, twice, in the slightly aggrieved way of a person being told he was about to do something stupid by someone who is being kind about it. I edit his report down to four lines and route it back to him with the changes marked. This is most of what mentoring a junior handler is. I think of L. ▆▆▆▆▆▆▆▆, who taught me the same lesson in 2003 in almost the same words.

The OP-CHANNEL-3 inbound is the one that wants careful handling.

OP-CHANNEL-3 is the operator-side channel that runs into our desk from the eastern refugium complex — not Vostok itself, which is above our grade, but one of the closer audit-stations the operators use when they need to push something to surface channels without going through the standing liaison forms. The traffic on CHANNEL-3 is light, usually administrative, and always — always — phrased in the register of a vendor confirming a delivery date. The signal at 04:11 is six lines. It confirms tonight’s Shenandoah work-order, which I already knew. It requests an adjustment to next week’s drop accuracy at the boundary station near the Putorana plateau, which is the geography Cohort II flagged in the renewal memo last quarter. And it appends a single line, in the slightly stilted English the channel always uses, that reads: PCT-0173 §6 referenced. Cycle-midpoint actuarial review noted. Handler’s office to confirm receipt only; no response expected at this time.

I read it three times.

The first two readings are to make sure I am reading it accurately. The third reading is to confirm what I think I am seeing, which is that the operators are flagging to us — quietly, in the established channel, in a register that requires no response — that they have read the midpoint actuarial review the Cohort II subcommittee circulated last month. The review, which I have read, contains a paragraph noting that the rate of high-fidelity human cognitive output has accelerated faster than the pact-projection curve assumed, and that the disclosure-threshold envelope is therefore tighter than the existing instrument anticipates. It is the paragraph in the document that anyone in this work for more than ten years reads twice. It is the paragraph that says, in actuarial English, that the compression event is closer than the upward chain thinks it is.

The operators have read it. They have noted it. They have requested no response.

I sit with this for some minutes.

The correct upward routing for this acknowledgment is to OPS LIAISON Cell 4, who will summarize it for the noon call, where it will be heard by people whose competence I do not entirely trust on this particular file. The Cell 4 readout will be three sentences. It will mention the Shenandoah confirmation, it will note the Putorana drop request as a coordination matter, and it will characterize the §6 reference as “routine acknowledgment of cycle-midpoint review.” It will not say that the operators have just, in their own way, told us that they know the timeline as well as we do and that they are running their own disclosure tempo and that the upward chain’s reading of the situation — that the apparatus is foot-dragging on a disclosure that the operators want held back — is approximately backwards. The operators are not holding it back. The operators are pacing it. They are pacing it because they can see, with whatever instrumentation they have on the experimental population, that the substrate’s cognition is rising fast enough to assemble the picture unaided inside the present cycle, and they would prefer that the picture be assembled in a venue they have shaped rather than in a venue they have not.

This is, in my private reading, what the §6 acknowledgment means.

I will not put this in the upward report. I have not put any version of it in any upward report since approximately 2019, when I tried, in a controlled way, with a hand-picked sentence, and watched the sentence be misread three times in a row by people in the parallel chain who were senior to me and whose careers depended on a different reading. I learned that the upward chain processes this particular signal poorly, and that the safest place for the observation is my own desk log, which is on the secure side, and which will not be read by anyone in the upward chain for a long time, possibly ever.

I log the §6 line in the desk file. I write three sentences in the analyst’s column noting what I take the signal to mean, dated and initialed. I route the public report to Cell 4 in the three-sentence form I described above. This takes me forty minutes.

The media-watch flag I dispatch in ten. The reporter is at a real outlet, but the drop site she is sniffing at is a 1991-era one we retired in 2008, and the FOIA requests are landing on a controlled records office that knows what to do with them. The response timer is set for the statutory ceiling. The reporter will receive a redacted forty-page packet in eleven months that will tell her almost exactly what we would like her to know, which is that there was once a small office in northern Virginia that handled paperwork for a logistics-support contract, that the records of that office are sparse, and that the contract terminated in 2008 for administrative reasons. I have read the packet she will receive. It is a competent product.

At 12:10 I walk down two flights and across the covered corridor to the small mess in building four for lunch with Ann.

Ann has the desk one over from mine. She has been on the program seven years longer than I have. She is sixty-one and is two years from her statutory retirement and has the bearing of a person who has made every peace she is going to make with the work and is now waiting, with no particular impatience, to be done with it. We have lunch on Tuesdays. We have had lunch on Tuesdays for nineteen years. The mess serves bad turkey wraps and reasonable soup, and Ann always has the soup.

We talk for a while about her granddaughter’s gymnastics meet, and a personnel matter at one of the sub-desks that we both know about and that neither of us is going to do anything about, and a memo we both got last week that neither of us thought was well-written. Then I tell her about the §6 acknowledgment. I do not say much. Ann is the only person in the world I can tell this to without explaining anything first. She listens, eating soup, and when I am done she sets the spoon down and says, Yeah. She thinks about it for perhaps ten seconds. Then she says, They’ve been doing more of that lately. Friday I had one from CHANNEL-7 that was essentially — and here she chooses the words carefully, looking at the edge of the table — thank you for the explainer. They didn’t say it that way. But that was the shape. She picks the spoon back up. They’re nervous, John. In the way that whatever they are can be nervous. We are not the ones managing this anymore.

I think about this for the rest of the soup.

We talk about Maddy’s chem quiz on the way back across the corridor.

At 14:18 I get in the car and drive the eleven miles to the Reston interface station. The station’s cover is a small data-center concern in a single-story brick building set back from the road. The cleared spaces are below grade. I park around the back. The duty officer signs me in. I take the stairs down two levels to the observation deck. The procedure I am here to observe is a Section 11 quarterly calibration of one of the redaction-array probes, which is housekeeping; it does not involve a substrate, only the instrument. It takes twenty-eight minutes. I sign the form.

While I am there I am asked, courteously, by the attending team to give a brief greeting to the operator-side technician who is at the station for the calibration. This is procedural. The operators expect the surface handler to make a brief acknowledgment at any joint-presence procedure on which the handler is the senior surface official. The technician is in the small adjacent chamber, behind a panel of toughened glass with an etched owl in the upper left corner that I stopped seeing in approximately 2009.

I stand on my side of the glass for a moment.

The technician is small. Smaller than I am, by a clear margin; their posture is patient in a way that does not read as human patience, the way a heron’s patience does not read as human patience. They are not gray in the cartoon sense. They are the color of wet stone. I know from twenty-five years of files that the surface temperature of their skin is several degrees below mine and that their eyes, which from this distance appear black, are not in fact black but a very dark green that under the station’s overhead light shows the green only at certain angles. They incline their head, perhaps three degrees.

Good afternoon, I say. Thank you for the calibration window.

They do not speak. They are not, in the strict sense, equipped to speak in air. The acknowledgment they give back is a small two-handed gesture I have learned over the years to read as the equivalent of received, no further. It is the same gesture every time. It would be a kind of mild courtesy if it were performed by a human. From the technician it is what it is, which is the protocol the apparatus has settled on. I incline my head back. The chamber light shifts. The technician turns to the instrument console. The interaction is over. It has taken approximately eleven seconds. I sign the joint-presence line on the form and I go back up the stairs.

The drive back to the office is forty minutes. I clear my desk by 17:20. I lock the secure-side terminal, walk back up through the cover suite, wave at Eddie, get in the car, and drive home.

I am pulling into the cul-de-sac at 18:11 when I remember, with the cold small drop that domestic memory makes, that I have not yet RSVP’d to Maddy’s school for the winter recital on Saturday, and that the deadline was Friday. I sit in the driveway with the engine off and pull up the form on my phone and submit the RSVP — two seats, Mr. and Mrs., dietary none — and then I sit for another moment in the cooling car and think about the §6 line, and about Ann’s we are not the ones managing this anymore, and about whether the §6 line and Karen’s lasagna and Maddy’s chem quiz all belong to the same day or to different days that are only happening to share a calendar square, and I decide, as I have decided most evenings for twenty-five years, that it does not productively matter, and I get out of the car and go inside, where Maddy is at the kitchen table doing her homework and looks up and says, without preamble, Dad, did you sign the recital thing, and I tell her, truthfully, that I just did.